Supercomputing Frontiers: 7th Asian Conference, SCFA 2022, Singapore, March 1–3, 2022, Proceedings (Lecture Notes in Computer Science)

This open access book constitutes the refereed proceedings of the 7 th Asian Conference Supercomputing Conference, SCFA 2022, which took place in Singapore in March 2022. The 8 full papers presented in this book were carefully reviewed and selected from 21 submissions. They cover a range of topics including file systems, memory hierarchy, HPC cloud platform, container image configuration workflow, large-scale applications, and scheduling. Preface Organization Contents High Performance Parallel LOBPCG Method for Large Hamiltonian Derived from Hubbard Model on Multi-GPU Systems 1 Introduction 2 LOBPCG Method for Solving Multi Eigenpairs on Multi-GPU Systems 2.1 Hamiltonian Matrix-Vector Multiplications 2.2 Vector Operations 3 Optimization CPU-GPU Data Transfer 3.1 Asynchronous Data Transfer 3.2 Reduction of Data Transfers 4 Numerical Experiments 5 Conclusions References Vapor Condensation Under Electric Field: A Study Using Molecular Dynamics Simulation 1 Introduction 2 Model and Methods 3 Results and Discussion 4 Conclusion References The Effect of Wing Mass and Wing Elevation Motion During Insect Forward Flight 1 Introduction 2 Methodology 2.1 Governing Equations and Numerical Discretization 2.2 Projection Method 2.3 SVD-GFD Scheme 2.4 Computation Acceleration 3 Results and Discussion 4 Conclusion References Exploring the Dynamics of Quantum Information in Many-Body Localised Systems with High Performance Computing 1 Introduction 2 Background and Numerical Setup for Dynamics 2.1 Physical Model 2.2 Simulation of Unitary Dynamics 3 Information Scrambling and Delocalization in MBL Systems 3.1 Channel-State Duality 3.2 Temporal Mutual Information 3.3 Problem Statement 3.4 Numerical Results 4 Conclusion and Discussion References On the Difference Between Shared Memory and Shared Address Space in HPC Communication 1 Introduction 2 Background and Related Work 3 Process-in-Process (PiP) 4 Shared Memory Vs. Shared Address Space 4.1 Page Tables and Page Faults 4.2 Modifications to Page Tables 5 Evaluation 5.1 Experimental Environment 5.2 Intel MPI Benchmark (IMB) Performance 5.3 Mini App Performance 6 Discussion 7 Summary References Evaluating GPU Programming Models for the LUMI Supercomputer 1 Introduction 2 Related Work 3 Programming Models 3.1 HIP 3.2 The OpenMP Application Programming Interface 3.3 SYCL 3.4 OpenACC 3.5 Alpaka 3.6 Kokkos 4 Choosing a Programming Model 5 Benchmarks and Applications 5.1 BabelStream 5.2 MiniBUDE 6 Methodology 6.1 Compilation 6.2 Execution and Tuning 7 Results 7.1 Configuration 7.2 BabelStream 7.3 MiniBUDE 8 Conclusion and Future Work References Evaluating Methods of Transferring Large Datasets 1 The Outline of the Problem 2 Tests Workbench: Hardware Specification 2.1 Servers 2.2 Mellanox InfiniBand Range Extenders 3 Storage Benchmark - Elbencho 3.1 The Motivation for Storage Benchmarking 3.2 Lots of Small Files Problem (LOSF) 3.3 The Reasons for Conducting ``Storage Sweeps'' 3.4 Storage Benchmarks Results 4 Network Benchmarks and Data Transfer Tests 4.1 Tests Methodology 4.2 Results 4.3 Additional Comments on the Tests Results 5 Conclusions 5.1 Future Work References Service Function Chaining Design & Implementation Using Network Service Mesh in Kubernetes 1 Introduction 2 Background 2.1 Motivation 2.2 Kubernetes 2.3 Containers 2.4 Different CNI Plug-ins 2.5 Network Service Mesh (NSM) 3 Design 4 Implementation 5 Performance Evaluation 6 Limitations and Future Work 7 Related Work 8 Conclusion References Author Index This open access book constitutes the refereed proceedings of the 6th Asian Supercomputing Conference, SCFA 2020, which was planned to be held in February 2020, but unfortunately, the physical conference was cancelled due to the COVID-19 pandemic. The 8 full papers presented in this book were carefully reviewed and selected from 22 submissions. They cover a range of topics including file systems, memory hierarchy, HPC cloud platform, container image configuration workflow, large-scale applications, and scheduling.